
A developer is told the lender goes to 70% LVR. The end value is $10m. The developer models a $7m facility, sets equity accordingly, and signs a land contract. The offer comes back at $5.6m.
Nothing went wrong. The facility was sized on the other test.
Construction and development facilities are almost never capped by loan-to-value ratio alone. They are capped at the lesser of:
Typical settings on non-bank development facilities: 70% LVR and 80% of total development cost, applied simultaneously. Some lenders express it as 66.5% LVR against roughly 79.8% TDC. The exact numbers move; the structure does not.
GRV ex-GST of $10.0m, total development cost of $7.0m:
The facility is $5.6m. Equity required is $1.4m, not the nil the LVR test implied.
Now raise the cost. Same $10.0m GRV, but $8.0m of cost:
Still the cost test. It only stops binding once cost exceeds roughly 87.5% of end value — which is to say, once the project has almost no margin and is unfundable for a different reason.
On any well-margined development, the cost test binds. Which makes the LVR number you were quoted, in practice, the less relevant of the two.
Because it measures the developer’s skin in the game rather than the market’s optimism. LVR against end value is a forecast; cost is a fact. A lender capping at 80% of cost is requiring the sponsor to fund 20% of the project’s actual spend, regardless of how the completed valuation lands.
This has a counter-intuitive consequence. A higher end valuation does not increase your facility once the cost test is binding. Chasing an extra $500,000 on the as-if-complete valuation changes nothing if you are capped at 80% of a $7m cost base.
Because the cap is a percentage of cost, understating cost shrinks your own facility. On one application the loan-to-cost ratio came back at 87.21% — over the 80% ceiling. Credit’s requested fix was not to reduce the loan. It was to substantiate additional legitimate project costs to bring the ratio under 80%.
Costs commonly left out of a first-pass feasibility, all of which properly belong in total development cost:
This is not creative accounting. It is putting real costs in the model so the ratio reflects the real project.
A loan-term mismatch materially understates required equity. Correcting one model from an assumed 12-month term to the actual 18 months — a 15-month build plus buffer — reduced the senior facility’s contribution to site value from $393,110 to $237,679, and lifted the equity required accordingly.
Confirm the modelled term matches the real program before you show a client a number. Many lenders default an indicative quote to a placeholder 24-month term until a QS report lands, then reset the formal term to QS-confirmed build period plus a buffer. Some also require the loan term to run at least six months longer than the build term as a structural rule.
An apparently inflated term often decomposes sensibly: an 18-month quote as 6 months settlement lag + 10 months build + 2 months buffer.
Where the margin scheme applies, LVR is assessed against net realisable value ex-GST, not the GST-inclusive headline. On a facility secured against townhouses presold at a combined GST-inclusive $3,095,000, the funder backed the GST out first and assessed LVR on roughly $2.8m. Using the gross figure understates true LVR — and the lender will correct it and tighten the loan.
Always confirm whether land price, GRV and construction cost inputs are GST-inclusive or exclusive. An unstated mismatch can flip a marginal project’s return outright.
Beyond LVR and LTC, many development lenders carry a minimum profit-margin covenant — commonly around 15% margin on cost. A deal can pass both leverage tests and still fail here. The finance assessment in our development feasibility calculator tests LVR, an 80% loan-to-cost cap and a 15% margin covenant together, so you can see which one binds before you sign a land contract.
Related benchmarks worth knowing:
Mezzanine debt at 18–20% rarely rescues an undersized project. When return on cost is negative or marginal, the fix is yield, sale price, build cost or land price — not the funding stack. One four-townhouse site modelled at −18.5% to −32% return on cost was turned to +18.07% by amalgamating with an adjoining site and adding two more dwellings, spreading largely fixed costs over more saleable area.
General information only. It is not credit assistance, financial product advice, or an offer of finance, and it does not take account of your objectives, financial situation or needs. Ratios and figures are indicative observations over 2024–2026 and vary by lender and deal.